{"ID":5938069,"CreatedAt":"2026-07-07T03:14:33.014478982Z","UpdatedAt":"2026-07-08T01:00:23.257252134Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2607.04068","arxiv_id":"2607.04068","title":"UniSGR: Unified Framework for Semantic ID Generation and Ranking","abstract":"Recommendation systems play a pivotal role in modern e-commerce platforms. While generative retrieval has emerged as a promising paradigm for alleviating the limitations of multi-stage cascade architectures, existing methods still struggle with fine-grained multi-objective ranking. To bridge this gap, we propose UniSGR, a Unified framework for Semantic ID Generation and Ranking. UniSGR adopts a two-stage training paradigm: a multi-scenario pre-training stage that learns from mixed business-scenario data, followed by a scenario-specific alignment stage that jointly optimizes Value-Aware Parallel Multi-Token Prediction (VA-PMTP) and a unified multi-objective ranking module. To better align generation with downstream ranking, we introduce Task-Aware Tokens (TAT) guided by Funnel-Aware Contrastive Learning. Furthermore, we propose Semantic Tree Attention with Reorganized KV cache (STARK), an inference strategy that removes key efficiency bottlenecks in conventional beam search. Extensive offline experiments on a large-scale e-commerce platform demonstrate the effectiveness and scalability of UniSGR.","short_abstract":"Recommendation systems play a pivotal role in modern e-commerce platforms. While generative retrieval has emerged as a promising paradigm for alleviating the limitations of multi-stage cascade architectures, existing methods still struggle with fine-grained multi-objective ranking. To bridge this gap, we propose UniSGR...","url_abs":"https://arxiv.org/abs/2607.04068","url_pdf":"https://arxiv.org/pdf/2607.04068v1","authors":"[\"Jiawei Sun\",\"Jun Yang\",\"Ziyue Guo\",\"Dongyue Xu\",\"Jianan Yan\",\"Lifang Deng\",\"Xiaoyi Zeng\"]","published":"2026-07-05T01:00:58Z","proceeding":"cs.IR","tasks":"[\"cs.IR\"]","methods":"[\"Generative Adversarial Network\"]","has_code":false}